Dennis Paulson wrote:
>I'd like to know more details about the U. of Toronto research on
>jaegers, Alvaro. I assume it's molecular. Recent mitochondrial DNA work
>on phalaropes seemed to indicate that Wilson's isn't particularly close
>to Red and Red-necked, i.e., they evolved independently from different
>groups of sandpipers. If that's the case, and if the Pomarine is really
>more closely related to the skuas than to the Parasitic & Long-tailed,
>systematics based on assessments of morphology and general biology almost
>sound valueless. However, I guess I'm saying there will be an uphill
>battle to convince me of either of these findings.


Well, there is in fact a good reason *not* to assume that "molecular"
systematics is inherently superior to "morphological" systematics: the
method of *analysis* of the data is more important than the nature of the
characters. Both kinds of characters can be analyzed using cladistic
methods (based on shared possession of evolutionary novelties; this is the
modern, Good way to do it); both can be analyzed using phenetic methods
(based on overall similarity; this is the old, Bad way). Most analyses of
morphological characters done in the last 10 - 15 years have been
cladistic (though for historical/social reasons, ornithology has been slow
to adopt cladistics). Many analyses of molecular data have been phenetic,
in that they rely on overall similarity. Sibley's DNA-DNA hybridization
is an example of phenetic analysis; it is a distance measure based on
overall similarity of DNA. To further complicate things, mtDNA data can
take the form of nucleotide sequences, restriction site maps, or
restriction fragments. Restriction fragments were used in a lot of the
early mtDNA analyses, and are still being used in some analyses; cladistic
interpretation of fragment data is suspect (see paper by Swofford & Olsen
in Hillis & Moritz, 1990, _Molecular systematics_). Cladistic analyses of
morphological characters and of nucleotide sequences have similar degrees
of internal consistency: neither class of character can be judged superior
on the basis of actual track records. To some (myself included), the best
phylogenetic treatment includes cladistic analysis of both molecular and
morphological characters (and ideally including fossil as well as living
taxa). Incidentally, cladistic approaches to fine-scale relationships
among populations will be showing us some interesting things about
speciation in the near future.


I'd be interested in taking a look at the phalarope study; is it
published?


David Wright
dwright at u.washington.edu